Breast Milk vs Sucrose in Relieving Procedural Pain in Preterm Neonates During Blood Draw by Automated Heel Lance

Blood draw by automated heel lancet is a common painful procedure in the Neonatal Intensive care units (NICU). The cutaneous sensory receptors on all mucosal and cutaneous surfaces are developed by 20 weeks gestational age (GA) and the neuronal circuits that carry pain stimuli from skin to cerebral cortex are developed by around 26 weeks of GA. Recurrent pain in preterm neonates impacts their postnatal growth, development, is associated with altered pain perception, and increased behavioral responses for future painful procedures.

Oral sucrose is a commonly administered and effective non-pharmacologic intervention used in the NICU for pain relief. There is insufficient evidence on its mechanism of action, minimum effective dose, the effect of repeated doses on long-term neurodevelopmental outcomes, and the consistency in providing pain relief with repeated doses over time. Moreover, the analgesic effect of sucrose is a developmentally transient phenomenon. Another non-pharmacologic intervention studied for pain relief in neonates is breast milk. The analgesic property of breast milk is presumed to be from its sweet taste from lactose, flavor, and odor. Breast milk has high levels of tryptophan, a melatonin precursor which may increase secretion of beta-endorphin, an endogenous opioid.

There is a paucity of data comparing analgesic properties of breast milk and sucrose in preterm neonates, along with insufficient evidence of sucrose as mentioned earlier, The investigator conducted this prospective randomized controlled trial (RCT) comparing sucrose with breast milk for pain relief in preterm neonates.

The primary objective of the study is to compare pain scores between two groups (24% sucrose and breast milk) during blood draw by automated heel lancet, in preterm neonates. The pain scores between the two groups will also be compared at 30 seconds intervals until two minutes post-procedure.

The current study is conducted at Hutzel Women's Hospital and Children's Hospital of Michigan NICU. Institutional Review Board approval was obtained from Wayne State University prior to study initiation. Before the study was initiated, physicians were trained in assessing the PIPP-R pain scale. The pain scores were given by two physicians who were blinded to the treatment arm, at baseline, during, 30-, 60-, 90-, and 120 seconds post-procedure. These scores were given either at the time of the procedure or a video is recorded to be scored later.

Eligible participants were identified and written informed consent was obtained from the parents. The timing and need for a blood draw are decided by the clinical team and are performed by the nurse taking care of the participant. Randomization was done by the pharmacist through a computer-generated program. Participants were randomly assigned to receive either 0.5 ml 24 % sucrose or 2 ml breastmilk, 2 minutes prior to the procedure. On the day of the procedure, the pharmacist sent a closed sealed envelope with a randomization group to the nurse taking care of the participant. Pain scores are given by two physicians during and post-procedure.

Te Investigators collected maternal characteristics of participants like age, prenatal care, race, history of diabetes and hypertension, receipt of antenatal steroids, magnesium, histological diagnosis of chorioamnionitis, mode of delivery, and duration of rupture of membranes. Neonatal characteristics collected include sex, APGAR score at 1 and 5 minutes, birth weight, gestational age, weight on the day of the procedure, postmenstrual age, number of skin pricks prior to the day of the procedure, and adverse events pre-defined as heart rate >240beats/minute or heart rate <80 beats/minute for >20 seconds, oxygen saturation <80 for >20 seconds, no spontaneous respiration for >20 seconds, choking, gagging were collected.

For this non-inferiority trial, the investigators chose the clinically significant difference to be 2 between the median PIPP-R scores during the procedure The investigators required 88 patients (44 in each group) to show that the lower limit of one-sided 95 % CI will be above the non-inferiority limit of -2 for an alpha error of 5% and power of 90 %.